CN111429472A - Image recognition method and device, electronic equipment and storage medium - Google Patents
Image recognition method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111429472A CN111429472A CN202010184712.9A CN202010184712A CN111429472A CN 111429472 A CN111429472 A CN 111429472A CN 202010184712 A CN202010184712 A CN 202010184712A CN 111429472 A CN111429472 A CN 111429472A
- Authority
- CN
- China
- Prior art keywords
- pixel unit
- value
- image
- values
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 230000015654 memory Effects 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 12
- 230000007613 environmental effect Effects 0.000 abstract description 9
- 230000008859 change Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/593—Recognising seat occupancy
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The application provides an image identification method, an image identification device, electronic equipment and a storage medium, and relates to the technical field of image identification, wherein the method comprises the following steps: the method comprises the steps of acquiring corresponding sample values for all pixel units in a first acquired image, wherein the sample values are used for indicating values of corresponding pixel units in a second acquired image before the first image, comparing the values of the pixel units with the corresponding sample values, determining that the corresponding pixel units belong to background pixel units if the difference values are smaller than set threshold values of the corresponding pixel units, updating the set threshold values of the corresponding background pixel units according to the difference values, respectively setting the threshold values for different pixel units, and continuously updating the corresponding threshold values according to the values of the corresponding pixel units when the pixel units are determined to belong to the background, so that the threshold values can adapt to environmental changes, false alarm is avoided, the identification accuracy is high, and the technical problems that in the prior art, foreground and background in the image are identified by adopting a fixed threshold value mode, and the misidentification rate is high are solved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to the field of image recognition technologies, and in particular, to an image recognition method and apparatus, an electronic device, and a storage medium.
Background
In recent years, with the popularization of motor vehicles, there are increasing incidents of forgetting children to death in a closed compartment after a driver stops and leaves the vehicle, resulting in immeasurable losses. Therefore, it is necessary to automatically monitor whether there are moving objects in the car, including children and pets, and to alarm and inform the car owner in time to avoid harm.
In the prior art, a technology of detecting a moving object in a scene based on a static camera is usually adopted, and whether the object in an image is a dynamic object of a foreground or a fixed background is identified through a set fixed threshold, while an identification mode of the fixed threshold cannot adapt to a scene with changed background light, and under the condition that the background light changes, the background is easily identified as a moving foreground object by mistake, so that the false alarm rate is high.
Disclosure of Invention
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first objective of the present application is to provide an image recognition method, so as to set corresponding thresholds for different pixel units, and continuously update the corresponding thresholds according to a value of the pixel unit when it is determined that the pixel unit belongs to a background, so that the thresholds can adapt to environmental changes, false alarms are avoided, and accuracy of recognition is improved.
A second object of the present application is to provide an image recognition apparatus.
A third object of the present application is to propose a computer device.
A fourth object of the present application is to propose a non-transitory computer-readable storage medium.
In order to achieve the above object, an embodiment of a first aspect of the present application provides an image recognition method, including:
acquiring a first image;
acquiring corresponding sample values for each pixel unit in the first image; the sample value is used for indicating the value of a corresponding pixel unit in a second image acquired before the first image;
comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold value of the corresponding pixel unit, determining that the corresponding pixel unit belongs to the background pixel unit;
and updating the set threshold of the corresponding background pixel unit according to the difference.
In order to achieve the above object, a second aspect of the present application provides an image recognition apparatus, including:
the acquisition module is used for acquiring a first image;
the acquisition module is used for acquiring corresponding sample values for each pixel unit in the first image; the sample value is used for indicating the value of a corresponding pixel unit in a second image acquired before the first image;
the comparison module is used for comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold value of the corresponding pixel unit, the corresponding pixel unit is determined to belong to the background pixel unit;
and the updating module is used for updating the set threshold of the corresponding background pixel unit according to the difference value.
To achieve the above object, an embodiment of a third aspect of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the image recognition method according to the first aspect.
In order to implement the foregoing embodiments, a fourth aspect of the present application proposes a non-transitory computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the image recognition method according to the first aspect.
The technical scheme provided by the embodiment of the application can have the following beneficial effects:
the method comprises the steps of collecting a first image, obtaining corresponding sample values for all pixel units in the first image, wherein the sample values are used for indicating values of corresponding pixel units in a second image collected before the first image, comparing the values of all the pixel units with the corresponding sample values, determining that the corresponding pixel units belong to background pixel units if the difference values are smaller than set threshold values of the corresponding pixel units, updating the set threshold values of the corresponding background pixel units according to the difference values to realize that the threshold values are respectively set for different pixel units, and continuously updating the corresponding threshold values according to the value conditions of the pixel units under the condition that the pixel units are determined to belong to the background, so that the threshold values can adapt to environmental changes, false alarm is avoided, and the identification accuracy is improved.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flowchart of an image recognition method according to an embodiment of the present disclosure;
fig. 2 is a second schematic flowchart of an image recognition method according to an embodiment of the present application;
fig. 3 is a third schematic flowchart of an image recognition method according to an embodiment of the present application;
fig. 4 is a fourth schematic flowchart of an image recognition method according to an embodiment of the present application;
fig. 5 is a fifth flowchart illustrating an image recognition method according to an embodiment of the present application;
fig. 6 is a sixth schematic flowchart of an image recognition method according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an image recognition apparatus according to an embodiment of the present disclosure; and
fig. 8 is a schematic diagram of an electronic device and a readable storage medium provided in the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
An image recognition method, an apparatus, an electronic device, and a storage medium according to embodiments of the present application are described below with reference to the drawings.
In the prior art, moving objects in an image are detected by adopting a Gaussian Mixture Model (GMM) method and a Visual background extraction (ViBe) algorithm, and the like, and moving objects in a monitoring scene of a fixed background collected by a camera are identified, wherein a fixed motionless region is identified as a background, and a moving object region is identified as a foreground, however, both the methods are based on a fixed threshold mode, and in practical application scenes, the complexity of the scenes is different, such as underground parking lots, open parking lots, trees and the like, and the scene has complex light changes of dim light, bright light, strong light irradiation when other vehicles pass through, tree shadow shaking and the like, and by adopting a fixed threshold mode, false detection is easy to occur, and meanwhile, the fixed threshold is not easy to set accurately, is small in setting, and is easy to occur false detection, the arrangement is large, and the omission is easy.
In order to solve the problem, embodiments of the present application provide an image recognition method, which sets corresponding thresholds in different regions of an image, and when it is determined that a pixel unit belongs to a background, the corresponding thresholds are continuously updated according to values of the corresponding pixel units, so that the thresholds can adapt to environmental changes, false alarms are avoided, and accuracy of recognizing a dynamic object in the image is improved.
Fig. 1 is a schematic flow chart of an image recognition method according to an embodiment of the present disclosure.
As shown in fig. 1, the method comprises the steps of:
And acquiring a first image by using a camera, wherein the first image is a picture to be subjected to dynamic object identification in the picture.
102, for each pixel unit in the first image, obtaining a corresponding sample value, where the sample value is used to indicate a value of a corresponding pixel unit in a second image acquired before the first image.
The method in this embodiment is mainly applied to a static scene, such as an in-vehicle scene, where the position of a camera used for acquiring an image is fixed, and a picture acquired by the camera includes an acquired current image to be recognized and also includes an image acquired before the current image to be recognized as a reference.
The pixel unit may include one pixel in an image or may include a plurality of pixels. The value of the pixel unit may be a gray value of the pixel unit.
Specifically, for each pixel unit in the first image, the value of the corresponding pixel unit corresponding to the position is obtained from the second image, and the sample value is initialized according to the obtained value of the corresponding pixel unit.
It should be understood that each pixel unit in the first image and the corresponding pixel unit in the second image refer to the pixel units with one-to-one correspondence positions in the image.
And 103, comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold of the corresponding pixel unit, determining that the corresponding pixel unit belongs to the background pixel unit.
Each pixel unit is provided with a corresponding set threshold, different set thresholds are used for different pixel units in the image, different areas in the image can be effectively resisted, and accurate determination of threshold setting is improved due to different illumination changes and noise degree interference.
Specifically, difference operation is performed on each pixel unit in the first image and the corresponding sample value to obtain a corresponding difference value, and if the difference value is smaller than a set threshold, it is determined that the corresponding pixel unit in the first image belongs to the background pixel unit.
And 104, updating the set threshold of the corresponding background pixel unit according to the difference value.
Specifically, for each background pixel unit, the difference value obtained by operation is used as an updated value, the updated value and the set threshold value before updating are weighted according to the set learning rate to obtain the updated set threshold value of the corresponding background pixel unit, so that the dynamic change of the threshold value along with the shot image is realized, and the threshold value can adapt to the environmental change by continuously updating the threshold value, and the false alarm is avoided.
For example, the pixel cell a and the corresponding sample value are calculated to obtain the corresponding difference value, the difference value is used as the update value S', and the threshold before update is set as S1The updated threshold is S2If the set learning rate is a, the updated threshold value S2Can be formulated as: s2＝(1-a)*S1+ a + S'. The learning rate is preferably, for example, 0.001, though it is small.
In the image identification method of the embodiment of the application, corresponding sample values are obtained for each pixel unit in a collected first image, wherein the sample values are used for indicating the value of a corresponding pixel unit in a second image collected before the first image, the value of each pixel unit is compared with the corresponding sample value to obtain a difference value, if the difference value is smaller than a set threshold value of the corresponding pixel unit, the corresponding pixel unit is determined to belong to a background pixel unit, the set threshold value of the corresponding background pixel unit is updated according to the difference value, the threshold values are respectively set for different pixel units, and when the pixel unit is determined to belong to the background, the corresponding threshold value is continuously updated according to the value of the corresponding pixel unit, so that the threshold value can adapt to environmental changes, false alarm is avoided, and the identification accuracy is high.
Based on the previous embodiment, this embodiment provides another image identification method, which increases the number of sample values to improve the accuracy of identifying whether the corresponding pixel unit in the first image belongs to the background pixel unit.
Fig. 2 is a second flowchart of an image recognition method according to an embodiment of the present application.
As shown in fig. 2, this may include the steps of:
The first image is a currently acquired frame image, and the second image is a frame image acquired before the first image.
In this embodiment, in order to further improve the accuracy of predicting whether each pixel unit in the first image belongs to the background, the value of the corresponding pixel unit is obtained from the previously acquired second image, and the values of the multiple adjacent pixel units of the corresponding pixel unit are obtained from the second image.
Specifically, the values of a plurality of adjacent corresponding pixel units in the second image are sequentially obtained according to the obtained value of the corresponding pixel unit in the second image and the distance between the corresponding pixel unit and the corresponding pixel unit, and a plurality of sample values are obtained after initialization.
It should be understood that each pixel in the first image, and the corresponding pixel in the second image, refers to the pixel with one-to-one corresponding position in the image, in order to improve the accuracy of each pixel cell identification, not only the values of the corresponding pixel cell in the second image are used for initializing the sample values, but also the values of a plurality of neighboring pixel cells of the corresponding pixel cell in the second image are used for initializing the sample values, for example, the number of the plurality of adjacent pixel units of the corresponding pixel unit in the second image is 20, thereby enlarging the corresponding sample value from a corresponding one of the pixel units in the second image to 21 pixel units, enlarging the range of the sample values, to realize that the corresponding pixel unit, and the value fluctuation range of each pixel unit in the first image is predicted by the values of the adjacent pixel units, so that the accuracy of predicting whether each pixel unit belongs to the background is improved.
And 204, comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold of the corresponding pixel unit, determining that the corresponding pixel unit belongs to the background pixel unit.
As a possible implementation manner, after sample values are initialized, a preset number of sample values are selected from the corresponding sample values, wherein the preset number of sample values is less than or equal to the total number of the sample values, preferably, the preset number is the total number of the sample values, further, a standard deviation corresponding to the corresponding pixel unit is determined by calculating the preset number of sample values, and 3 times of the value of the standard deviation is used as the set threshold, so that different set thresholds are used for different pixel units in the first image, different areas in the image can be effectively resisted, interference caused by different illumination changes and noise degrees is avoided, and accurate determination of threshold setting is improved.
It should be understood that, the 3-time threshold is determined according to the 3sigma principle of gaussian distribution, and 99.73% of the pixels are included in the 3sigma range, that is, the accuracy of the identification can be as high as 99.73, so that the set threshold is set, when whether the corresponding pixel unit of the first image is the background pixel unit is identified, the first image is compact and not prone to false alarm, and the set threshold is smaller and easier to false alarm, and larger and easier to miss detection.
And then, determining whether the corresponding pixel unit belongs to a background pixel unit based on a set threshold corresponding to the corresponding pixel unit, specifically, performing difference operation on each pixel unit in the first image and the corresponding multiple sample values to obtain a corresponding difference value, and if at least the difference value of the set number is smaller than the set threshold, determining that the corresponding pixel unit in the first image belongs to the background pixel unit.
For example, a pixel cell a in the first image compares difference values with corresponding 21 sample values, and if at least 2 difference values are smaller than a set threshold of the pixel cell, it is determined that the corresponding pixel cell in the first image belongs to a background pixel cell.
Specifically, for each background pixel unit, an update value is determined according to a difference obtained by operation, as a possible implementation manner, if at least two differences smaller than a set threshold are obtained, an average value of the at least two differences is used as the update value; as another possible implementation manner, if the difference value smaller than the set threshold is one, the difference value is used as an updated value, the updated value and the set threshold before updating are weighted according to the set learning rate to obtain the updated set threshold of the corresponding background pixel unit, so that the dynamic change of the threshold along with the shot image is realized, and the threshold is continuously updated to adapt to the environmental change, thereby avoiding false alarm.
For example, the pixel unit a and the corresponding 21 sample values are respectively calculated to obtain corresponding difference values, and if the difference values are different in size, as a possible implementation manner, the difference values may be calculated to obtain an average value, the average value is used as an updated value S ', and if the difference values are the same, the difference value is used as an updated value S', and the threshold before updating is set to be S1The updated threshold is S2If the set learning rate is a, the updated threshold value S2Can be formulated as: s2＝(1-a)*S1+ a + S'. The learning rate is preferably, for example, 0.001, though it is small.
In the image identification method of this embodiment, when the sample values are generated, in addition to obtaining the values of the corresponding pixel units from the second image, the values of a plurality of adjacent pixel units of the corresponding pixel units are also obtained, so as to increase the number of related sample values, and improve the accuracy of predicting whether each pixel unit in the first image belongs to the background.
Based on the previous embodiment, this embodiment provides a possible implementation manner of the image recognition method, and provides a method that can quickly determine a difference between a value of each pixel unit in the first image and a corresponding sample value, so as to quickly update the set threshold corresponding to each pixel unit in the first image.
Fig. 3 is a third schematic flowchart of an image recognition method according to an embodiment of the present application.
Specifically, the steps 301 to 303 can specifically refer to the steps 201 and 203 in the previous embodiment, and the principle is the same, which is not described herein again.
Specifically, according to the value of the corresponding pixel unit in the acquired second image and the distance between the acquired value and the corresponding pixel unit, the values of a plurality of adjacent corresponding pixel units are acquired, a plurality of sample values are obtained after initialization, the corresponding pixel unit is taken as a reference, the first H sample values closest to the corresponding pixel unit are taken as a head set, and the rest sample values are taken as a non-head set. For example, the value of the corresponding pixel unit a in the second image is determined, the values of the N nearest neighboring pixel units are sequentially taken according to the distance from the pixel unit a, and the sample value of the corresponding pixel unit in the first image is obtained through initialization. The head set includes sample values corresponding to the first H pixel units closest to the pixel unit a, with the pixel unit a as a reference, and the sample values corresponding to the remaining adjacent pixel units are the sample values of the non-head set.
In this embodiment, to increase the comparison speed, each pixel unit in the first image is compared with each sample value in the corresponding header set.
Specifically, according to the samples in the header set, the calculated difference is compared with the set threshold corresponding to the corresponding pixel unit in the first image, the number of sample values with the difference smaller than the set threshold is counted, and according to the number of sample values with the difference smaller than the set threshold, whether difference comparison is continued or whether the corresponding pixel unit in the first image belongs to the background is determined, and the difference comparison with the sample values can be stopped.
And 308, when the number of the sample values with the difference value smaller than the set threshold value is not smaller than the set number, determining that the corresponding pixel unit belongs to the background pixel unit.
Specifically, if the number of the difference values smaller than the set threshold is smaller than the set number, it cannot be determined whether the corresponding pixel unit in the first image belongs to the background pixel unit, the value of the corresponding pixel unit is continuously compared with the sample values in the corresponding non-header set, the number of the difference values smaller than the set threshold is continuously counted and accumulated, and when the number of the difference values smaller than the set threshold satisfies the set number, it is determined that the corresponding pixel unit in the first image belongs to the background pixel unit.
Specifically, if the difference calculated between the corresponding pixel unit in the first image and the sample in the corresponding header set is smaller than the number of the set threshold, and not smaller than the set number, it may already be determined that the corresponding pixel unit in the first image belongs to the background pixel unit, so that the comparison of the difference between the sample values in the corresponding non-header set may be stopped, the speed of comparing the difference is increased, and it is implemented to quickly determine whether the corresponding pixel unit in the first image belongs to the background pixel unit.
Specifically, reference may be made to step 205 in the previous embodiment, which has the same principle and is not described herein again.
In the image identification method of this embodiment, the sample values are divided into the header set with a higher correlation degree and the non-header set with a relatively low correlation degree, the corresponding pixel unit of the first image is compared with each sample value in the header set by the difference value, it is determined whether the number of sample values in the header set, for which the difference value is smaller than the set threshold value, is not smaller than the set number, if not, the difference value comparison is stopped, and it is determined that the corresponding pixel unit of the first image belongs to the background pixel unit, so that the speed of identifying whether the corresponding pixel unit of the first image is the background pixel unit is increased, and the efficiency is improved.
Based on the foregoing embodiments, this embodiment provides still another image identification method, after determining that a corresponding pixel unit belongs to a background pixel unit, the sample value is further updated, so that the sample value is also correspondingly updated with each frame of the captured image, so that the sample value is also more adaptive to changes in environment and light, and such changes can be extended to adjacent pixel units.
Fig. 4 is a fourth schematic flowchart of the image recognition method according to the embodiment of the present application, and as shown in fig. 4, after determining that the corresponding pixel unit belongs to the background pixel unit, the method may further include the following steps:
Specifically, after it is determined that the corresponding pixel unit of the first image belongs to the background pixel unit, the corresponding pixel unit in the first image is the background pixel unit, and according to the value of the background pixel unit, one of the multiple sample values corresponding to the background pixel unit is updated, that is, one of the multiple corresponding sample values is randomly selected for updating, so as to maintain the dynamic update of the sample values, so that even if the illumination is slowly changed, for example, the intensity of sunlight changes over time, false alarm cannot be caused because the sample values are fixed and unchanged.
In this embodiment, after the sample values corresponding to the corresponding pixel units in the second image corresponding to the corresponding pixel units in the first image are randomly updated, one of the multiple sample values corresponding to the adjacent pixel units may be updated according to the value of the background pixel unit, that is, the multiple sample values corresponding to the adjacent pixel units are randomly updated, so as to implement spreading of the update of the sample to the multiple sample values corresponding to the adjacent pixel units.
In the previous embodiment, after it is determined that the corresponding pixel unit belongs to the background pixel unit, the update is performed by using any one of the plurality of sample values corresponding to the corresponding pixel unit, so as to ensure the continuous update of the sample library. In this embodiment, how to exchange pixel units in the header set and the non-header set according to a plurality of sample values corresponding to the background pixel unit after determining that the corresponding pixel unit in the first image belongs to the background pixel unit is described, so that samples similar to the corresponding background pixel unit are divided into the header set, the number of comparison times of each pixel unit in the first image is reduced, and the speed of identifying the background pixel unit is increased.
Fig. 5 is a fifth flowchart illustrating an image recognition method according to an embodiment of the present application.
As shown in fig. 5, after determining that the corresponding pixel unit belongs to the background pixel unit, the following steps may be further included:
And the value difference between the target sample value and the corresponding background pixel unit is smaller than a set threshold value.
Specifically, after determining that the corresponding pixel unit in the first image belongs to the background pixel unit, the corresponding pixel unit of the first image is determined as the background pixel unit, the value of the corresponding pixel unit of the first image and the corresponding multiple sample values are subjected to difference value operation, and the corresponding sample value when the difference value is smaller than the set threshold value is determined as the target sample value.
Specifically, since the samples have been previously divided into the header set and the non-header set in the present application, and the target sample is still one of the multiple samples, it may be determined whether the target sample belongs to the header set, and if the target sample value is not in the corresponding header set, the target sample value is exchanged with any one of the sample values in the header set, so that the difference between the updated sample value in the header set and the value of the background pixel unit is smaller and smaller than the difference between the sample value in the non-header set and the value of the background pixel unit, so that the header region may help to improve the speed of identifying the background pixel unit.
As a possible implementation manner, according to corresponding pixel units of a plurality of frames of a first image, a plurality of sample values included in a corresponding header set are sequentially replaced, so that a plurality of sample values in the header set are updated frame by frame along with the first image. For example, the header set includes 4 sample values, which are named sample 1, sample 2, sample 3 and sample 4, then the first image of the first frame updates sample 1 in the header set, the first image of the second frame updates sample 2 in the header set, the first image of the third frame updates sample 3 in the header set, the first image of the fourth frame updates sample 4 in the header set, and the subsequent frames are updated from sample 1 again, so that the sample values in the header set can be dynamically updated to adapt to illumination changes, thereby improving reliability of the sample values in the corresponding header set.
It should be noted that if more than one target sample value is determined, any one of the target sample values is swapped with one of the sample values in the header set.
In the image identification method in the embodiment of the application, for a background pixel unit, a target sample value is determined from a plurality of corresponding sample values, where a value difference between the target sample value and the corresponding background pixel unit is smaller than a set threshold, and if the target sample value is not in a corresponding head set, the target sample value is exchanged with one sample value in the head set, so that it is ensured that a difference between the value of the sample value in the head set and the value of the current background pixel unit is smaller than a difference between the value of the sample value in a non-head set and the value of the background pixel unit, so that the head area can help to improve the speed of identifying the background pixel unit.
Based on the foregoing embodiment, a possible implementation manner of the image recognition method provided in this embodiment further illustrates how to determine the foreground region in the first image, and fig. 6 is a sixth schematic flowchart of the image recognition method provided in this embodiment of the present application.
As shown in fig. 6, after determining that the corresponding pixel unit belongs to the background pixel unit, the following steps may be further included:
And determining the confidence degree according to the times that the corresponding pixel unit in the second image acquired before the first image does not belong to the background pixel unit.
For example, the confidence of the pixel unit in the acquired image is in a range of [0,100], that is, the initial value of the confidence is 0, and each increase is increased by 10 according to a fixed interval, wherein the confidence threshold is set to be 30. If the pixel unit a in the currently acquired first image frame is determined not to belong to the background pixel unit, the confidence level of the pixel unit a is increased from an initial value 0 to 40, the confidence level value of the pixel unit a is greater than a confidence level threshold value 30, and the pixel unit a is determined to belong to the foreground pixel unit, that is, not to belong to the background pixel unit. Further, as the number of frames acquired increases, the threshold of pixel cell a does not increase when it increases to a maximum value of 100. If the pixel unit a is determined to belong to the background pixel unit by connecting 7 frames with the acquired frame number further increased, the confidence value 100 of the pixel unit a is reduced to 30, and the pixel unit a is determined to belong to the background pixel unit again, that is, the pixel unit a is regressed from the pixel unit belonging to the foreground to the pixel unit belonging to the background pixel unit, and the accuracy of determining the background pixel unit is improved through the dynamic change of the confidence.
In this embodiment, in the first image, according to each pixel unit whose confidence is greater than the confidence threshold, a plurality of maximum connected regions are generated, where the maximum connected regions include a connected region including only a small number of pixel units, and the connected regions including only a small number of sporadic pixel units can be removed by setting an area threshold, and the maximum connected region whose area is greater than the set area threshold is used as a foreground region in the first image, so as to improve accuracy of determining the foreground region.
In the image identification method in this embodiment, the confidence is added to the remaining pixel units in the first image that do not belong to the background pixel unit, the maximum connected region is generated for each pixel unit in the first image whose confidence is greater than the confidence threshold, and the maximum connected region whose area is greater than the set area threshold is used as the foreground region in the first image, so that the removal of the suspected foreground region is realized, and the accuracy of determining the foreground region is improved.
In order to implement the above embodiments, the present application further provides an image recognition apparatus.
Fig. 7 is a schematic structural diagram of an image recognition apparatus according to an embodiment of the present application.
As shown in fig. 7, the apparatus includes: an acquisition module 61, an acquisition module 62, a comparison module 63 and an update module 64.
The acquisition module 61 is configured to acquire a first image.
The obtaining module 62 is configured to obtain a corresponding sample value for each pixel unit in the first image, where the sample value is used to indicate a value of a corresponding pixel unit in a second image acquired before the first image.
And a comparing module 63, configured to compare the value of each pixel unit with the corresponding sample value, and determine that the corresponding pixel unit belongs to the background pixel unit if the difference is smaller than the set threshold of the corresponding pixel unit.
And an updating module 64, configured to update the set threshold of the corresponding background pixel unit according to the difference.
Further, in a possible implementation manner of the embodiment of the present application, the apparatus further includes: the device comprises a switching module, an adding module and a determining module.
The switching module is used for determining a target sample value from a plurality of corresponding sample values for the background pixel unit; the value difference between the target sample value and the corresponding background pixel unit is smaller than a set threshold value; and if the target sample value is not in the corresponding head set, exchanging the target sample value with one sample value in the head set.
The increasing module is used for increasing confidence coefficient of the rest pixel units which do not belong to the background pixel unit in the first image; and determining the confidence coefficient according to the times that the corresponding pixel unit in the second image acquired before the first image does not belong to the background pixel unit.
The determining module is used for generating a maximum connected region for each pixel unit with the confidence coefficient larger than a confidence threshold value in the first image; and taking the maximum connected region with the area larger than the set area threshold value as a foreground region in the first image.
As a possible implementation manner, the update module 64 includes:
and the determining unit is used for determining an updating value for each background pixel unit according to the difference value.
And the calculating unit is used for weighting the updated value and the set threshold before updating according to the set learning rate to obtain the updated set threshold of the corresponding background pixel unit.
As a possible implementation manner, if at least two difference values smaller than the set threshold are provided, the determining unit is specifically configured to:
taking an average of at least two of the differences as the update value.
As a possible implementation manner, the obtaining module 62 includes:
an obtaining unit, configured to obtain, for each pixel unit in the first image, a value of the corresponding pixel unit from the second image, and obtain values of a plurality of adjacent pixel units of the corresponding pixel unit from the second image;
and the initialization unit is used for initializing a plurality of sample values according to the acquired values of the corresponding pixel units and the values of the adjacent pixel units.
As a possible implementation manner, the update module 64 is further configured to:
and updating one of the corresponding sample values according to the value of the background pixel unit.
As a possible implementation manner, the update module 64 is further configured to:
and updating one of a plurality of sample values corresponding to the adjacent pixel units according to the value of the background pixel unit.
As a possible implementation manner, the comparing module 63 is specifically configured to:
comparing the difference value with a plurality of corresponding sample values for each pixel unit in the first image; and if the difference value of at least the set number is smaller than the set threshold value, determining that the corresponding pixel unit in the first image belongs to the background pixel unit.
As a possible implementation manner, the obtaining module 62 further includes:
a dividing unit for dividing the plurality of sample values into a header set and a non-header set according to distances between pixel units of the initialized plurality of sample values;
thus, the comparing module 63 is further configured to:
comparing a difference value between each pixel unit in the first image and each sample value in the corresponding head set, determining the number of sample values in the head set, wherein the difference value is smaller than a set threshold value, and if the determined number is smaller than the set number, comparing the value of the corresponding pixel unit with each sample value in the corresponding non-head set; and if the determined number is not less than the set number, stopping comparing the difference value with each sample value in the corresponding non-head set.
It should be noted that the foregoing explanation of the embodiment of the image recognition method is also applicable to the image recognition apparatus of this embodiment, and is not repeated here.
In the image recognition device according to the embodiment of the application, corresponding sample values are obtained for each pixel unit in a first collected image, wherein the sample values are used for indicating the value of a corresponding pixel unit in a second collected image before the first image, the value of each pixel unit is compared with the corresponding sample value to obtain a difference value, if the difference value is smaller than a set threshold value of the corresponding pixel unit, the corresponding pixel unit is determined to belong to a background pixel unit, the set threshold value of the corresponding background pixel unit is updated according to the difference value, the threshold values are respectively set for different pixel units, and when the pixel unit is determined to belong to the background, the corresponding threshold value is continuously updated according to the value of the corresponding pixel unit, so that the threshold value can adapt to environmental changes, false alarm is avoided, and the recognition accuracy is high.
In order to implement the foregoing embodiments, an electronic device is provided in an embodiment of the present application, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the electronic device implements the image recognition method according to the foregoing method embodiments.
In order to implement the above embodiments, the present application provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the computer program implements the image recognition method according to the foregoing method embodiments.
According to an embodiment of the present application, an electronic device and a readable storage medium are also provided.
As shown in fig. 8, the electronic device is a block diagram of an electronic device according to an image recognition method of an embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.
As shown in fig. 8, the electronic apparatus includes: one or more processors 1001, memory 1002, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions for execution within the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input/output apparatus (such as a display device coupled to the interface). In other embodiments, multiple processors and/or multiple buses may be used, along with multiple memories and multiple memories, as desired. Also, multiple electronic devices may be connected, with each device providing portions of the necessary operations (e.g., as a server array, a group of blade servers, or a multi-processor system). Fig. 8 illustrates an example of a processor 1001.
The memory 1002 is a non-transitory computer readable storage medium provided herein. The memory stores instructions executable by at least one processor to cause the at least one processor to perform the image recognition method provided herein. The non-transitory computer-readable storage medium of the present application stores computer instructions for causing a computer to execute the recognition method of an image provided by the present application.
The memory 1002, as a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the image recognition method in the embodiment of the present application (for example, the acquisition module 61, the acquisition module 62, the comparison module 63, and the update module 64 shown in fig. 7). The processor 1001 executes various functional applications of the server and data processing, i.e., implements the recognition method of the image in the above-described method embodiment, by running non-transitory software programs, instructions, and modules stored in the memory 1002.
The memory 1002 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the electronic device of the recognition method of the image, and the like. Further, the memory 1002 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 1002 may optionally include a memory remotely located from the processor 1001, and such remote memory may be connected to the electronic device of the image recognition method through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The electronic device of the image recognition method may further include: an input device 1003 and an output device 1004. The processor 1001, the memory 1002, the input device 1003, and the output device 1004 may be connected by a bus or other means, and the bus connection is exemplified in fig. 8.
The input device 1003 may receive input numeric or character information and generate key signal inputs related to user settings and function controls of the electronic device of the method of recognition of the image, such as a touch screen, keypad, mouse, track pad, touch pad, pointing stick, one or more mouse buttons, track ball, joystick, etc. the output device 1004 may include a display device, auxiliary lighting device (e.g., L ED), and tactile feedback device (e.g., vibrating motor), etc.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, programmable logic devices (P L D)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal.
The systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or L CD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer for providing interaction with the user.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., AN application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with AN implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
According to the technical scheme of the embodiment of the application, a first image is collected, corresponding sample values are obtained for all pixel units in the first image, wherein the sample values are used for indicating values of corresponding pixel units in a second image collected before the first image, the values of all the pixel units are compared with the corresponding sample values, if the difference value is smaller than a set threshold value of the corresponding pixel unit, the corresponding pixel unit is determined to belong to a background pixel unit, the set threshold value of the corresponding background pixel unit is updated according to the difference value, so that the threshold values are respectively set for different pixel units, and the corresponding threshold values are continuously updated according to the value conditions of the pixel units under the condition that the pixel units are determined to belong to the background, so that the threshold values can adapt to environmental changes, false alarm is avoided, and the identification accuracy is improved.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present application may be executed in parallel, sequentially, or in different orders, and the present invention is not limited thereto as long as the desired results of the technical solutions disclosed in the present application can be achieved.
The above-described embodiments should not be construed as limiting the scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (22)
1. A method for recognizing an image, the method comprising:
acquiring a first image;
acquiring corresponding sample values for each pixel unit in the first image; the sample value is used for indicating the value of a corresponding pixel unit in a second image acquired before the first image;
comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold value of the corresponding pixel unit, determining that the corresponding pixel unit belongs to the background pixel unit;
and updating the set threshold of the corresponding background pixel unit according to the difference.
2. The method for recognizing an image according to claim 1, wherein the updating the set threshold of the corresponding background pixel unit according to the difference value comprises:
for each background pixel unit, determining an updated value according to the difference value;
and weighting the updated value and the set threshold before updating according to a set learning rate to obtain the updated set threshold of the corresponding background pixel unit.
3. The method according to claim 2, wherein if there are at least two differences smaller than the set threshold, the determining an updated value according to the differences comprises:
taking an average of at least two of the differences as the update value.
4. The method according to claim 1, wherein the obtaining corresponding sample values for each pixel unit in the first image comprises:
for each pixel unit in the first image, obtaining the value of the corresponding pixel unit from the second image, and obtaining the values of a plurality of adjacent pixel units of the corresponding pixel unit from the second image;
and initializing a plurality of sample values according to the acquired values of the corresponding pixel units and the values of the adjacent pixel units.
5. The image recognition method according to claim 4, wherein the comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold of the corresponding pixel unit, determining that the corresponding pixel unit belongs to the background pixel unit, further comprises:
and updating one of the corresponding sample values according to the value of the background pixel unit.
6. The method according to claim 4, wherein the comparing the value of each pixel unit with the corresponding sample value is performed by comparing a difference value, and if the difference value is smaller than the set threshold of the corresponding pixel unit, the corresponding pixel unit is determined to belong to a background pixel unit, and the method further comprises
And updating one of a plurality of sample values corresponding to the adjacent pixel units according to the value of the background pixel unit.
7. The method according to claim 4, wherein comparing the value of each pixel unit with the corresponding sample value to determine that the corresponding pixel unit belongs to the background pixel unit if the difference is smaller than the set threshold of the corresponding pixel unit comprises:
comparing a difference value with a plurality of corresponding sample values for each pixel unit in the first image;
and if the difference value of at least the set number is smaller than the set threshold value, determining that the corresponding pixel unit in the first image belongs to the background pixel unit.
8. The image recognition method according to claim 4, wherein after initializing the plurality of sample values according to the obtained values of the corresponding pixel unit and the values of the plurality of adjacent pixel units, the method further includes:
dividing the plurality of sample values into a header set and a non-header set according to the distance between the initialized pixel units of the plurality of sample values, and comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, including:
comparing a difference value with each sample value in the corresponding head set for each pixel unit in the first image;
determining the number of sample values in the header set, wherein the difference value is smaller than the set threshold;
if the determined number is smaller than the set number, comparing the value of the corresponding pixel unit with each sample value in the corresponding non-head set;
and if the determined number is not less than the set number, stopping comparing the difference value with each sample value in the corresponding non-head set.
9. The method of claim 8, wherein after determining that the corresponding pixel unit belongs to the background pixel unit, the method further comprises:
for the background pixel unit, determining a target sample value from a plurality of corresponding sample values; wherein, the value difference between the target sample value and the corresponding background pixel unit is less than the set threshold;
and if the target sample value is not in the corresponding head set, exchanging the target sample value with one sample value in the head set.
10. The method for recognizing an image according to any one of claims 1 to 9, wherein after determining that the corresponding pixel unit belongs to the background pixel unit, the method further comprises:
increasing confidence coefficient for the rest pixel units in the first image, which do not belong to the background pixel unit; the confidence coefficient is determined according to the number of times that the corresponding pixel unit in the second image acquired before the first image does not belong to the background pixel unit;
generating a maximum connected region for each pixel unit of the first image, wherein the confidence coefficient is greater than a confidence threshold value;
and taking the maximum connected region with the area larger than a set area threshold value as a foreground region in the first image.
11. An apparatus for recognizing an image, the apparatus comprising:
the acquisition module is used for acquiring a first image;
the acquisition module is used for acquiring corresponding sample values for each pixel unit in the first image; the sample value is used for indicating the value of a corresponding pixel unit in a second image acquired before the first image;
the comparison module is used for comparing the value of each pixel unit with the corresponding sample value to obtain a difference value, and if the difference value is smaller than the set threshold value of the corresponding pixel unit, the corresponding pixel unit is determined to belong to the background pixel unit;
and the updating module is used for updating the set threshold of the corresponding background pixel unit according to the difference value.
12. The image recognition device of claim 11, wherein the update module comprises:
a determining unit, configured to determine, for each background pixel unit, an update value according to the difference;
and the calculating unit is used for weighting the updated value and the set threshold before updating according to a set learning rate to obtain the updated set threshold of the corresponding background pixel unit.
13. The image recognition device of claim 12, wherein if there are at least two difference values smaller than the set threshold, the determining unit is specifically configured to:
taking an average of at least two of the differences as the update value.
14. The image recognition device of claim 11, wherein the acquisition module comprises:
an obtaining unit, configured to obtain, for each pixel unit in the first image, a value of the corresponding pixel unit from the second image, and obtain values of a plurality of adjacent pixel units of the corresponding pixel unit from the second image;
and the initialization unit is used for initializing a plurality of sample values according to the acquired values of the corresponding pixel units and the values of the adjacent pixel units.
15. The image recognition device of claim 14, wherein the update module is further configured to:
and updating one of the corresponding sample values according to the value of the background pixel unit.
16. The image recognition device of claim 14, wherein the update module is further configured to:
and updating one of a plurality of sample values corresponding to the adjacent pixel units according to the value of the background pixel unit.
17. The image recognition device of claim 14, wherein the comparison module is specifically configured to:
comparing a difference value with a plurality of corresponding sample values for each pixel unit in the first image;
and if the difference value of at least the set number is smaller than the set threshold value, determining that the corresponding pixel unit in the first image belongs to the background pixel unit.
18. The image recognition device of claim 14, wherein the acquisition module further comprises:
a dividing unit for dividing the plurality of sample values into a header set and a non-header set according to distances between pixel units of the initialized plurality of sample values;
the comparison module is further configured to:
comparing a difference value with each sample value in the corresponding head set for each pixel unit in the first image;
determining the number of sample values in the header set, wherein the difference value is smaller than the set threshold;
if the determined number is smaller than the set number, comparing the value of the corresponding pixel unit with each sample value in the corresponding non-head set;
and if the determined number is not less than the set number, stopping comparing the difference value with each sample value in the corresponding non-head set.
19. The image recognition device of claim 18, further comprising:
the switching module is used for determining a target sample value from a plurality of corresponding sample values for the background pixel unit; wherein, the value difference between the target sample value and the corresponding background pixel unit is less than the set threshold; and if the target sample value is not in the corresponding head set, exchanging the target sample value with one sample value in the head set.
20. The image recognition device according to any one of claims 11 to 19, further comprising:
an increasing module, configured to increase confidence for remaining pixel units in the first image that do not belong to the background pixel unit; the confidence coefficient is determined according to the number of times that the corresponding pixel unit in the second image acquired before the first image does not belong to the background pixel unit;
the determining module is used for generating a maximum connected region for each pixel unit of which the confidence coefficient is greater than a confidence threshold value in the first image; and taking the maximum connected region with the area larger than a set area threshold value as a foreground region in the first image.
21. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of image recognition according to any one of claims 1 to 10 when executing the program.
22. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method of recognizing an image according to any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010184712.9A CN111429472B (en) | 2020-03-16 | 2020-03-16 | Image recognition method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010184712.9A CN111429472B (en) | 2020-03-16 | 2020-03-16 | Image recognition method, device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111429472A true CN111429472A (en) | 2020-07-17 |
| CN111429472B CN111429472B (en) | 2023-11-03 |
Family
ID=71547995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010184712.9A Active CN111429472B (en) | 2020-03-16 | 2020-03-16 | Image recognition method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111429472B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101533511A (en) * | 2009-04-14 | 2009-09-16 | 清华大学 | Background image updating method and device thereof |
| CN103236160A (en) * | 2013-04-07 | 2013-08-07 | 水木路拓科技(北京)有限公司 | Road network traffic condition monitoring system based on video image processing technology |
| CN104978734A (en) * | 2014-04-11 | 2015-10-14 | 北京数码视讯科技股份有限公司 | Foreground image extraction method and foreground image extraction device |
| CN106447674A (en) * | 2016-09-30 | 2017-02-22 | 北京大学深圳研究生院 | Video background removing method |
| US20170161905A1 (en) * | 2015-12-07 | 2017-06-08 | Avigilon Analytics Corporation | System and method for background and foreground segmentation |
| CN110114801A (en) * | 2017-01-23 | 2019-08-09 | 富士通株式会社 | Display foreground detection device and method, electronic equipment |
| EP3543902A1 (en) * | 2018-03-22 | 2019-09-25 | Canon Kabushiki Kaisha | Image processing apparatus and method and storage medium storing instructions |
| CN110610507A (en) * | 2018-06-14 | 2019-12-24 | 安讯士有限公司 | Method, device and system for determining whether pixel position belongs to background or foreground |
| CN110879948A (en) * | 2018-09-06 | 2020-03-13 | 华为技术有限公司 | Image processing method, device and storage medium |
-
2020
- 2020-03-16 CN CN202010184712.9A patent/CN111429472B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101533511A (en) * | 2009-04-14 | 2009-09-16 | 清华大学 | Background image updating method and device thereof |
| CN103236160A (en) * | 2013-04-07 | 2013-08-07 | 水木路拓科技(北京)有限公司 | Road network traffic condition monitoring system based on video image processing technology |
| CN104978734A (en) * | 2014-04-11 | 2015-10-14 | 北京数码视讯科技股份有限公司 | Foreground image extraction method and foreground image extraction device |
| US20170161905A1 (en) * | 2015-12-07 | 2017-06-08 | Avigilon Analytics Corporation | System and method for background and foreground segmentation |
| CN106447674A (en) * | 2016-09-30 | 2017-02-22 | 北京大学深圳研究生院 | Video background removing method |
| CN110114801A (en) * | 2017-01-23 | 2019-08-09 | 富士通株式会社 | Display foreground detection device and method, electronic equipment |
| EP3543902A1 (en) * | 2018-03-22 | 2019-09-25 | Canon Kabushiki Kaisha | Image processing apparatus and method and storage medium storing instructions |
| CN110610507A (en) * | 2018-06-14 | 2019-12-24 | 安讯士有限公司 | Method, device and system for determining whether pixel position belongs to background or foreground |
| CN110879948A (en) * | 2018-09-06 | 2020-03-13 | 华为技术有限公司 | Image processing method, device and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| JIN-BIN YANG 等: "A New Method for Motion Target Detection by Background Subtraction and Update" * |
| 张挺 等: "基于改进 Jung 方法的实时背景提取算法" * |
| 杨大勇 等: "基于动态阈值的核密度估计前景检测算法" * |
| 陈凤东 等: "基于动态阈值背景差分算法的目标检测方法" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111429472B (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220383535A1 (en) | Object Tracking Method and Device, Electronic Device, and Computer-Readable Storage Medium | |
| Rakibe et al. | Background subtraction algorithm based human motion detection | |
| CN112507949A (en) | Target tracking method and device, road side equipment and cloud control platform | |
| US20150310297A1 (en) | Systems and methods for computer vision background estimation using foreground-aware statistical models | |
| CN111292531A (en) | Tracking method, device and equipment of traffic signal lamp and storage medium | |
| CN108961293B (en) | Background subtraction method, device, equipment and storage medium | |
| CN112785628A (en) | Track prediction method and system based on panoramic view angle detection and tracking | |
| CN115049954B (en) | Target identification method, device, electronic equipment and medium | |
| CN112989987B (en) | Method, apparatus, device and storage medium for identifying crowd behavior | |
| CN111950345B (en) | Camera identification method and device, electronic equipment and storage medium | |
| CN114091515A (en) | Obstacle detection method, obstacle detection device, electronic apparatus, and storage medium | |
| CN113989696A (en) | Target tracking method and device, electronic equipment and storage medium | |
| CN110728700B (en) | Moving target tracking method and device, computer equipment and storage medium | |
| CN111798356A (en) | Rail transit passenger flow abnormal pattern recognition method based on big data | |
| Ji et al. | Sparse regularized correlation filter for UAV object tracking with adaptive contextual learning and keyfilter selection | |
| Chen et al. | A unified model sharing framework for moving object detection | |
| CN107301655B (en) | Video moving target detection method based on background modeling | |
| CN110798681B (en) | Monitoring method and device of imaging equipment and computer equipment | |
| CN113065523A (en) | Target tracking method and device, electronic equipment and storage medium | |
| CN111401248A (en) | Sky area identification method and device, electronic equipment and storage medium | |
| Gao et al. | Moving object detection for video surveillance based on improved ViBe | |
| CN111429472A (en) | Image recognition method and device, electronic equipment and storage medium | |
| Srilekha et al. | A novel approach for detection and tracking of vehicles using Kalman filter | |
| CN114445711B (en) | Image detection method, image detection device, electronic equipment and storage medium | |
| CN113989695B (en) | Target tracking method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211014 Address after: 100176 101, floor 1, building 1, yard 7, Ruihe West 2nd Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing Applicant after: Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. Address before: 2 / F, baidu building, 10 Shangdi 10th Street, Haidian District, Beijing 100085 Applicant before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |